-8 DegC, car didn't start
 

Any ideas on why my car wouldn't start? I've attached a copy of my .msq, it starts in anything warmer than -8, but this morning no dice.

messing with the cranking enrichments seemed to have very little effect. If I added fuel it just kept turning over, but when I finally gave up cranking the engine backfired.

Take fuel out and this stopped happening. Car still didn't start mind.

I did eventually get it going as one of the 'backfires' kicked the car into life. :vash:

Starting hasn't been an issue for my car for goddamn years so I'm kinda pissed it's suddenly started becoming a problem.

I've attached a datalog of the non start (you can see the minor rpm bum just after I finish turning the key and my .msq (MSII 3.0.3u sequential fuel, Toyota COPS, 2001 VVT engine, 96 CAS).

There is a trick here. If you can't get it to start when its too cold, you can't tell if there is too much or not enough fuel. So the trick is to try to start and if it doesn't start, then try again in anti-flood mode (full throttle while cranking). If it starts or almost starts, then you have too much fuel and you need to remove some. Otherwise, you need more fuel.

It can get tricky with big injectors, as 1% is a significant amount of fuel.

I did try cranking a few times throttle fully depressed to clear anythoug out of the cylinders, but it just turned over.

To be honest the car never even gave a hint that it was going to start, sometimes you get encouragement from occasional firing, but I had nothing, zip, nada.

One thought has just occoured to me though, battery voltage was down to about 8.5V at times, could my dwell settings be way off? I have Toyota COPS, but not the usual ones, they're 12v triggered ones from (I think) Toyota VVT motors.

Would a hotter spark help?

Hrm, Reverant, CRAIGO very kindly sent me a copy of his E-DIYPNP .msq to have a look at.

Looking at his ignition, his cranking dwell is 6ms, whereas mine is 3. Also his running dwell is a good 5ms longer.

I pulled my dwell pretty much from my ass as I had nothing to go on for it and no way of measuring the correct dwell required. AFAIK Toyota COPS have overdwell protection built in, so *hopefully* I won't cook my coils attempting this tonight.

Before I do though would my lack of dwell be a major contributor to no cold start in cold weather? The car has been running fine (if a little on the poor side MPG wise) since the start of the year when I fitted the engine.

For Toyota cops use 2ms dwell running. Cranking still needs like 5-6ms. When running you should be closer to 2.4-2.5ms dwell, check your dwell gauge.

When using flood clear, watch your Pulse Width and make sure it's actually at Zero when cranking.

Why don't you just move somewhere warmer?

Are you offering to put me up Hustler? I'm sure I could put up with the rednecks and driving on the wrong side of the road for that bit of warmth.

Cheers Brain, I've upped my running dwell to 3ms and cranking to 6ms, the car started fine at lunchtime, however coolant had 'only' dropped to 10degC and IAT to -1 so not the same conditions.

I swear it felt smoother but placebo is a bitch.

Flood clear was definately cutting PW, I have another datalog much longer of unsucsseful starting and it shows flood clear working.

EDIT: checked my cranking dwell in a log and it's 8ms, can anyone explain that to me as cranking dwell was set at 3.5 ms, where's the other 4.5 ms come from?

voltage.

under Advanced, IIRC, there is a battery dwell voltage curve. Was this altered for the Toyota Cops? I use the curve as measured by Jason, and then for Max dwell duration I use 2.0ms, this put me closer to the ideal dwell value when running at 14v.

Right, still no dice here.

I've tried a replacement battery, and have also tickle charged my current one, no change, it's not a lack of juice issue.

Increasing cranking dwell has made no difference, (though running dwell has improved the engine smoothness a lot, especially at idle).

Any other ideas? What's really pissing me off is it starts perfectly at or above 0degC, it's not as if the curve looks whacked out below 0DegC either :(

If I add fuel it makes no difference, the engine just cranks and cranks, however as soon as I stop cranking it backfires, it's these backfires which actually start my car in this weather. If I pull 1-2% of fuel the backfires stop, but the car will just crank forever

WTF is going on?

P.S. Brain, what does your Dwell battery correction curve look like? mine is 500% at 6V, 250% at 8v, 170% at 10V 130% at 12V 102% at 14V and 88% at 16V, cheers :)

what's your cranking RPM?


The ideal dwell battery voltage curve for toyota cops is:

6.0v 228
8.0v 180
10v 148
12v 118
14v 100
16v 88

then set the maximum dwell duration to 2.1. This should result in a 2.1ms dwell when running at 14v, and you should see 2.5ms dwell at 12v.

Cranking RPM at that temp is 200-220~ not exactly high I know, but ISTR it being about that in warmer temps too, I'm not 100% on that however.

I'll try your dwell curve, much appreciated - although I'm not on the 'normal' Toyota COPS, so it may not be an exact fit. It could be that at the low voltages I'm seeing the COPS are entering failsafe and not sparking - grasping at straws though I feel :(

Also it's been suggested that sequential might be a cause, I'll try moving to untimed/batch tomorrow morning, just to see if it makes a difference.

What's really bugging me is the complete lack of an attempt to start, but as soon as you stop turning the key it backfires. Altering the cranking PW only changes whether this backfire occours or not.

what is your cranking RPM value set at?

300

see what bumping it to 400 does.

How would that help?

AFAIK the cranking rpm just tells the ECU when it's no longer cranking?

I'll certainly give it a try, just wondering what your thought process was.

cause my car did something similar when the cranking was set too low, at like 40*F or under it would fail boat. at 45*F and above it would start right up.

It would be too rich regardless of where i set the pulsewidth, and while it didnt catch on key off. it took like 20 seconds of cranking with the throttle WOT to start...then i realized that flood clear wasnt kicking in, cause my RPM value was too low.

Well I've changed my dwell correction* curve and upped cranking rpm to 400, it started after both changes

BUT it's 10deg warmer, so not a good example.

* I added a dwell gauge prior and noted my cranking dwell was around 3.1 despite cranking dwell being set at 6ms, after changing the curve it stayed at 3.1, more investigation required...

Quick update, I think we might be getting somewhere :)

Car cranked for a bit then chugged, then started, I think it just needs the fuel dialling in properly now.

Pretty sure the issue was the dwell correction curve meant the coils went into failsafe meaning no spark & no start, I'll have a play if it's cold enough this weekend swopping the cranking speed back to 300 to see if it was that and actually nail what seems to have fixed it.

Cheers for your help Brain :)

Another update.

Still having issues, but I think my dad has figure the problem out.

My coils (Toyota COPS) are not the 'normal' ones we use on our cars, they're a bit longer than normal and crucially are 12v triggered rather than 5v triggered.

In the cold weather my voltage drops dramatically during cranking, enough to mean that no-where near enough voltage gets to the coils to trigger them (though they're 12v triggers they will trigger at lower than 12v).

How the hell can I fix this? I'm using Joe's ignition driver circuit aka two transistors to invert the signal.

I need some way of bumping up the amount of juice my coils see during really cold weather cranking.

BTW my battery is perfectly fine, it'll crank the engine for 15-20 minutes before slowing in this weather.

I'm currently driving around with a booster battery in the boot so 1st start in the morning I can give the system a bit of a gee up.

Second battery for the coils that is isolated during cranking but charges after cranking?